R Notebook

# load("/Volumes/Research Project/Self-Anchoring/Combined/Combined/reportingAnalysesEnv.RData")

library(pacman)
p_load(tidyverse, ggdist, wesanderson, patchwork, brms, igraph, GGally, ggrepel, network, sna, here, performance, ggpubr)

params.UCLA <- arrow::read_parquet(here::here("Study 2/Cleaning/output/S_Logistic_1m_Oppose_Biasparams.UCLA.parquet"))
params.CSULA <- arrow::read_parquet(here::here("Study 2/Cleaning/output/S_Logistic_1m_Oppose_Biasparams.CSULA.parquet"))
params.NotUCR <- arrow::read_parquet(here::here("Study 2/Cleaning/output/S_Logistic_1m_Oppose_Biasparams.NotUCR.parquet"))
params2 <- rbind(
  cbind(params.UCLA,condition="UCLA"),
  cbind(params.CSULA,condition="CSU LA"),
  cbind(params.NotUCR,condition="Not UCR")
  )

params.Majority <- arrow::read_parquet(here::here("Study 3/Cleaning/output/S_Logistic_1m_Oppose_Biasparams.Majority.parquet"))
params.Minority <- arrow::read_parquet(here::here("Study 3/Cleaning/output/S_Logistic_1m_Oppose_Biasparams.Minority.parquet"))
params3 <- rbind(
  cbind(params.Majority,condition="Majority"),
  cbind(params.Minority,condition="Minority")
  )

# params3 <- rbind(
#   cbind(params.Majority),#,condition="Majority"),
#   cbind(params.Minority)#,condition="Minority")
#   )

load adjencency matrix for positive

#pos_adj <- read.csv("input/pos_adj.csv")
pos_adj <- read.csv(here("Combined/input/adjacencyMatrix_p.csv"))
colnames(pos_adj) <- 1:148
pos_adj <- as.matrix(pos_adj)
#pos_att <- read.csv("input/traitLevelAttribute_p.csv")
pos_att <- read.csv(here("Combined/input/allPosCents.csv"))
# generate graph object from the matrix
graph_pos <- graph_from_adjacency_matrix(pos_adj, mode = "directed", weighted=NULL)
V(graph_pos)$tNames <- pos_att$trait
pos_att$traitFriendly <- pos_att$trait
pos_att$traitFriendly[pos_att$traitFriendly!="Friendly"] <- ""
V(graph_pos)$FriendlyName <- pos_att$traitFriendly

Focus on “Friendly” - High outdegree centrality trait

FriendlyNet <- ego(graph_pos, order = 1, nodes = 46, mode = "all",
  mindist = 0)
FriendlyNet <- induced.subgraph(graph_pos, vids = unlist(FriendlyNet) )

set.seed(2)

egoNet <- ggnet2(FriendlyNet, size = "outdegree",  color = "dodgerblue",  palette = "Set2", mode = "fruchtermanreingold", directed=TRUE, arrow.size = 3, arrow.gap = 0.025, layout.exp = .35, alpha = .7, edge.alpha = .45) + 
  guides(color = FALSE, size = FALSE) + geom_text_repel(aes(label = V(FriendlyNet)$tNames), size = 3.5, force = .025, fontface = "bold")
egoNet

ggsave(egoNet, file = "/Volumes/Research Project/Self-Anchoring/Plots/outdegreeNet.png", units = "in", dpi = 300, width= 6, height = 4)

ggsave(egoNet, file = "/Volumes/Research Project/Self-Anchoring/Plots/outdegreeNet.tiff", units = "in", dpi = 300, width= 6, height = 4)

Feedback Community Colored

library(wesanderson)

graph_pos2 <- intergraph::asNetwork(graph_pos)
#graph_pos2 %v% "cluster" = as.character(V(graph_pos)$cluster)

set.seed(5)

p1 <- ggnet2(graph_pos2, size = "outdegree",  color = "dodgerblue", palette = "Dark2", mode = "kamadakawai", directed=TRUE, layout.exp = .65, alpha = .7, edge.alpha = .25) +
  guides(color = FALSE, size = FALSE)
p1

ggsave(p1, file = "/Volumes/Research Project/Self-Anchoring/Plots/colorNet.png", units = "in", dpi = 300, width= 6, height = 4)

ggsave(p1, file = "/Volumes/Research Project/Self-Anchoring/Plots/colorNet.tiff", units = "in", dpi = 300, width= 6, height = 4)

stack <- ggpubr::ggarrange(p1, egoNet, ncol=1, nrow=2)

ggsave(stack, file = "/Volumes/Research Project/Self-Anchoring/Plots/stackNetwork.png", dpi = 300, units = "in", width= 8, height = 10)

ggsave(stack, file = "/Volumes/Research Project/Self-Anchoring/Plots/stackNetwork.tiff", dpi = 300, units = "in", width= 8, height = 10)

Probability of Significance

library(bayestestR)
ps1 <- bayestestR::p_significance(WSRNOV.ING)
PS1.WSRNOV.ING <- plot(ps1) + jtools::theme_apa() + scale_y_discrete(labels = c("Similarity-to-Self\nX Novel","Novel","Similarity-to-Self")) + xlab(expression(beta)) + ylab("Fixed Effects") + geom_vline(xintercept = .18) + geom_vline(xintercept = -.18) + jtools::theme_apa() + ggtitle(NULL)

Scale for y is already present.
Adding another scale for y, which will replace the existing scale.

ggsave(PS1.WSRNOV.ING, file = "/Volumes/Research Project/Self-Anchoring/Plots/PSplot.png", dpi = 300, units = "in", width= 7.5, height =4.5)

ggsave(PS1.WSRNOV.ING, file = "/Volumes/Research Project/Self-Anchoring/Plots/PSplot.tiff", dpi = 300, units = "in", width= 7.5, height =4.5)

ps1 <- bayestestR::p_significance(WSRNOV.ING)
PS1.WSRNOV.ING <- plot(ps1) + jtools::theme_apa() + scale_y_discrete(labels = c("Similarity-to-Self\nX Novel","Novel","Similarity-to-Self")) + xlab(expression(beta)) + ylab("Fixed Effects") + geom_vline(xintercept = .18) + geom_vline(xintercept = -.18) + jtools::theme_apa() + ggtitle(NULL) + scale_fill_manual(values = c("red","red","blue"))

ps2 <- bayestestR::p_significance(WSRNOV.ING.2)
PS2.WSRNOV.ING <- plot(ps2) + jtools::theme_apa() + scale_y_discrete(labels = c("Similarity-to-Self\nX Novel","Novel","Similarity-to-Self")) + xlab(expression(beta)) + ylab("Fixed Effects") + geom_vline(xintercept = .18) + geom_vline(xintercept = -.18) + jtools::theme_apa() + ggtitle(NULL) + scale_fill_manual(values = c("red","blue","blue"))

ps3 <- bayestestR::p_significance(WSRNOV.ING.3)
PS3.WSRNOV.ING <- plot(ps3) + jtools::theme_apa() + scale_y_discrete(labels = c("Similarity-to-Self\nX Novel","Novel","Similarity-to-Self")) + xlab(expression(beta)) + ylab("Fixed Effects") + geom_vline(xintercept = .18) + geom_vline(xintercept = -.18) + jtools::theme_apa() + ggtitle(NULL) + scale_fill_manual(values = c("red","blue","blue"))

ggpubr::ggarrange(PS1.WSRNOV.ING, PS2.WSRNOV.ING, PS3.WSRNOV.ING, nrow=3, ncol=1)

ggsave("/Volumes/Research Project/Self-Anchoring/Plots/PS_Stack.png", height = 8, width = 9, units = "in")

ggsave("/Volumes/Research Project/Self-Anchoring/Plots/PS_Stack.tiff", height = 8, width = 9, units = "in")

Posterior Predictive Check

ING1PPC <- posterior_predictive_check(WSR.ING) + jtools::theme_apa()

Using 10 posterior draws for ppc type 'dens_overlay' by default.

ING2PPC <- posterior_predictive_check(WSR.ING.COND.2) + jtools::theme_apa()

Using 10 posterior draws for ppc type 'dens_overlay' by default.

ING3PPC <- posterior_predictive_check(WSR.ING.COND.3) + jtools::theme_apa()

Using 10 posterior draws for ppc type 'dens_overlay' by default.

SR1PPC <- posterior_predictive_check(MCS.SR) + jtools::theme_apa()

Using 10 posterior draws for ppc type 'dens_overlay' by default.

SR2PPC <- posterior_predictive_check(MCS.SR.COND.2) + jtools::theme_apa()

Using 10 posterior draws for ppc type 'dens_overlay' by default.

SR3PPC <- posterior_predictive_check(MCS.SR.COND.3) + jtools::theme_apa()

Using 10 posterior draws for ppc type 'dens_overlay' by default.

COMBPPC <- ggpubr::ggarrange(ING1PPC, ING2PPC, ING3PPC, SR1PPC, SR2PPC, SR3PPC, labels = c("A", "B", "C", "D", "E", "F"), nrow = 2, ncol =3, common.legend = T)

ggsave(COMBPPC, file = "/Volumes/Research Project/Self-Anchoring/Plots/PPCplot.png", dpi = 300, units = "in", width= 10, height = 9)

ggsave(COMBPPC, file = "/Volumes/Research Project/Self-Anchoring/Plots/PPCplot.tiff", dpi = 300, units = "in", width= 10, height = 9)

Study 1

Self-Evaluation Ingroup Classification

DES.ING.1.p <- conditional_effects(DES.ING)#, method = "posterior_linpred")
DES.ING.1.p_ <- DES.ING.1.p$`desirability.Z`
DES.ING.1.plot <-ggplot(DES.ING.1.p_, aes(effect1__, estimate__)) +  geom_line(aes()) + 
  geom_ribbon(aes(ymin=lower__, ymax=upper__), alpha=0.15) + 
  theme(
    legend.margin = margin(6, 6, 6, 6)
    ) + theme(axis.text=element_text(size=12),
        axis.title=element_text(size=12,face="bold")) + theme(legend.text = element_text(size=12)) + theme(panel.border = element_rect(colour = "black", fill = NA, size =1)) + theme(legend.title = element_blank()) + theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank(),
panel.background = element_blank(), axis.line = element_line(colour = "black")) +
  xlab("Desirability") + ylab("Ingroup Classification")
DES.ING.1.plot

Self-Evaluation Ingroup Classification

SR.ING.1.p <- conditional_effects(SELF.ING)#, method = "posterior_linpred")
SR.ING.1.p_ <- SR.ING.1.p$`selfResp.Z`
SR.ING.1.plot <-ggplot(SR.ING.1.p_, aes(effect1__, estimate__)) +  geom_line(aes()) + 
  geom_ribbon(aes(ymin=lower__, ymax=upper__), alpha=0.15) + 
  theme(
    legend.margin = margin(6, 6, 6, 6)
    ) + theme(axis.text=element_text(size=12),
        axis.title=element_text(size=12,face="bold")) + theme(legend.text = element_text(size=12)) + theme(panel.border = element_rect(colour = "black", fill = NA, size =1)) + theme(legend.title = element_blank()) + theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank(),
panel.background = element_blank(), axis.line = element_line(colour = "black")) +
  xlab("Self-Evaluation") + ylab("Ingroup Classification")
SR.ING.1.plot

Similarity-to-Self Ingroup Classification

WSR.ING.1.p <- conditional_effects(WSR.ING)#, method = "posterior_linpred")
WSR.ING.1.p_ <- WSR.ING.1.p$`WSR.Z`
WSR.ING.1.plot <-ggplot(WSR.ING.1.p_, aes(effect1__, estimate__)) +  geom_line(aes()) + 
  geom_ribbon(aes(ymin=lower__, ymax=upper__), alpha=0.15) + 
  theme(
    legend.margin = margin(6, 6, 6, 6)
    ) + theme(axis.text=element_text(size=12),
        axis.title=element_text(size=12,face="bold")) + theme(legend.text = element_text(size=12)) + theme(panel.border = element_rect(colour = "black", fill = NA, size =1)) + theme(legend.title = element_blank()) + theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank(),
panel.background = element_blank(), axis.line = element_line(colour = "black")) +
  xlab("Similarity-to-Self") + ylab("Ingroup Classification")
WSR.ING.1.plot

Combine Plots

COMB.ING.1.plot1 <- ggarrange(DES.ING.1.plot, SR.ING.1.plot, WSR.ING.1.plot, labels=c("A","B","C"), nrow=1)

Metacontrast Ratio

MCS.SR.1.p <- conditional_effects(MCS.SR)#, method = "posterior_linpred")

Warning: Predictions are treated as continuous variables in 'conditional_effects' by default which is likely invalid for ordinal families. Please set 'categorical' to TRUE.

MCS.SR.1.p_ <- MCS.SR.1.p$`metaContrastS.Z`
MCS.SR.1.MCRplot <-ggplot(MCS.SR.1.p_, aes(effect1__, estimate__)) +  geom_line(aes()) + 
  geom_ribbon(aes(ymin=lower__, ymax=upper__), alpha=0.15) +
  theme(axis.text=element_text(size=12),
        axis.title=element_text(size=12,face="bold")) + theme(legend.text = element_text(size=12)) + theme(panel.border = element_rect(colour = "black", fill = NA, size =1)) + theme(legend.title = element_blank()) + theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank(),
panel.background = element_blank(), axis.line = element_line(colour = "black")) +
  xlab("Metacontrast Ratio") + ylab("Self-Evaluation")
MCS.SR.1.MCRplot

Similarity-to-Ingroup

ISOS.SR.1.p <- conditional_effects(ISOS.SR)#, method = "posterior_linpred")

Warning: Predictions are treated as continuous variables in 'conditional_effects' by default which is likely invalid for ordinal families. Please set 'categorical' to TRUE.Warning: Predictions are treated as continuous variables in 'conditional_effects' by default which is likely invalid for ordinal families. Please set 'categorical' to TRUE.

ISOS.SR.1.p_ <- ISOS.SR.1.p$`inGsimS.Z`
ISOS.SR.1.InGplot <-ggplot(ISOS.SR.1.p_, aes(effect1__, estimate__)) +  geom_line(aes()) + 
  geom_ribbon(aes(ymin=lower__, ymax=upper__), alpha=0.15) +
  theme(axis.text=element_text(size=12),
        axis.title=element_text(size=12,face="bold")) + theme(legend.text = element_text(size=12)) + theme(panel.border = element_rect(colour = "black", fill = NA, size =1)) + theme(legend.title = element_blank()) + theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank(),
panel.background = element_blank(), axis.line = element_line(colour = "black")) +
  xlab("Similarity-to-Ingroup") + ylab("Self-Evaluation")
ISOS.SR.1.InGplot

Similarity-to-Outgroup

ISOS.SR.1.p <- conditional_effects(ISOS.SR)#, method = "posterior_linpred")

Warning: Predictions are treated as continuous variables in 'conditional_effects' by default which is likely invalid for ordinal families. Please set 'categorical' to TRUE.Warning: Predictions are treated as continuous variables in 'conditional_effects' by default which is likely invalid for ordinal families. Please set 'categorical' to TRUE.

ISOS.SR.1.p_ <- ISOS.SR.1.p$`outGsimS.Z`
ISOS.SR.1.OutGplot <-ggplot(ISOS.SR.1.p_, aes(effect1__, estimate__)) +  geom_line(aes()) + 
  geom_ribbon(aes(ymin=lower__, ymax=upper__), alpha=0.15) +
  theme(axis.text=element_text(size=12),
        axis.title=element_text(size=12,face="bold")) + theme(legend.text = element_text(size=12)) + theme(panel.border = element_rect(colour = "black", fill = NA, size =1)) + theme(legend.title = element_blank()) + theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank(),
panel.background = element_blank(), axis.line = element_line(colour = "black")) +
  xlab("Similarity-to-Outgroup") + ylab("Self-Evaluation")
ISOS.SR.1.OutGplot

Study 2

Warmth

warmthDf <- pivot_longer(indDiffs2, cols=starts_with("Therm_"), names_to="University", values_to="Warmth") %>% select(subID, University, Warmth) %>% drop_na()
warmthDf$University <- gsub("Therm_1","UCR", warmthDf$University)
warmthDf$University <- gsub("Therm_2","UCLA", warmthDf$University)
warmthDf$University <- gsub("Therm_4","CSU LA", warmthDf$University)

warmPlot2 <- ggplot(warmthDf, aes(University, Warmth)) + 
  ggdist::stat_halfeye(adjust = .5, width = .7, .width = 0, justification = -.2, point_colour = NA) + 
  geom_boxplot(width = .2, outlier.shape = NA)  + scale_x_discrete(labels=c("Lower Status\nOutgroup","Higher Status\nOutgroup", "Ingroup")) +
  geom_jitter(width = .05, alpha = .15) + labs(y="Perceived Warmth",x="University") + jtools::theme_apa()

Status

indDiffs2$Status <- NULL
statusDf <- pivot_longer(indDiffs2, cols=ends_with("_Status"), names_to="University", values_to="Status") %>% select(subID, University, Status) %>% drop_na()
statusDf$University <- gsub("_Status","", statusDf$University)

statusDf$University <- as.factor(statusDf$University)
statusDf$subID <- as.factor(statusDf$subID)

statusPlot2 <- ggplot(statusDf, aes(University, Status)) + 
  ggdist::stat_halfeye(adjust = .5, width = .7, .width = 0, justification = -.2, point_colour = NA) + 
  geom_boxplot(width = .2, outlier.shape = NA) + scale_x_discrete(labels=c("Lower Status\nOutgroup","Higher Status\nOutgroup", "Ingroup")) +
  geom_jitter(width = .05, alpha = .15) + labs(y="Perceived Status",x="University") + jtools::theme_apa()

Combine Warmth and Status Plots

wsPlot2 <- warmPlot2 + statusPlot2 

wsPlot2

Projection Rates

mDiffPlot2 <- ggplot(params2, aes(condition, m)) + 
  ggdist::stat_halfeye(adjust = .5, width = .7, .width = 0, justification = -.2, point_colour = NA) + 
  geom_boxplot(width = .2, outlier.shape = NA) + scale_x_discrete(labels=c("Lower Status","Negation", "Higher Status")) +
  geom_jitter(width = .05, alpha = .15) + labs(y="Projection Rate",x="University") + jtools::theme_apa()

Bias Parameter

bDiffPlot2 <- ggplot(params2, aes(condition, bias)) + 
  ggdist::stat_halfeye(adjust = .5, width = .7, .width = 0, justification = -.2, point_colour = NA) + 
  geom_boxplot(width = .2, outlier.shape = NA) + scale_x_discrete(labels=c("Lower Status","Negation", "Higher Status")) +
  geom_jitter(width = .05, alpha = .15) + labs(y="Bias Parameter",x="University") + jtools::theme_apa()

Combine Bias and Projection

mbPlot2 <- mDiffPlot2 + bDiffPlot2 

mbPlot2

Desirability Ingroup Classifications

DES.ING.COND.2.p <- conditional_effects(DES.ING.COND.2)#, method = "posterior_linpred")
DES.ING.COND.2.p_ <- DES.ING.COND.2.p$`desirability.Z:outgroup`
DES.ING.COND.2.plot <-ggplot(DES.ING.COND.2.p_, aes(effect1__, estimate__)) +  geom_line(aes(linetype=effect2__, color=effect2__)) + 
  geom_ribbon(aes(ymin=lower__, ymax=upper__, fill=effect2__), alpha=0.15) + 
  scale_linetype_discrete(labels = c("Negation","Lower Status","Higher Status")) + 
  scale_color_manual(labels = c("Negation","Lower Status","Higher Status"), values = wes_palette("Darjeeling1")) + 
  scale_fill_manual( 
                      labels=c("Negation","Lower Status","Higher Status"), values = wes_palette("Darjeeling1")) + 
  theme(
    legend.position = c(.08, .75),
    legend.justification = c("left", "bottom"),
    legend.box.just = "left",
    legend.margin = margin(6, 6, 6, 6)
    ) + theme(axis.text=element_text(size=12),
        axis.title=element_text(size=12,face="bold")) + theme(legend.text = element_text(size=12)) + theme(panel.border = element_rect(colour = "black", fill = NA, size =1)) + theme(legend.title = element_blank()) + theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank(),
panel.background = element_blank(), axis.line = element_line(colour = "black")) +
  xlab("Desirability") + ylab("Ingroup Classification")
DES.ING.COND.2.plot

Self-Evaluation Ingroup Classification

SR.ING.COND.2.p <- conditional_effects(SR.ING.COND.2)#, method = "posterior_linpred")
SR.ING.COND.2.p_ <- SR.ING.COND.2.p$`selfResp.Z:outgroup`
SR.ING.COND.2.plot <-ggplot(SR.ING.COND.2.p_, aes(effect1__, estimate__)) +  geom_line(aes(linetype=effect2__, color=effect2__)) + 
  geom_ribbon(aes(ymin=lower__, ymax=upper__, fill=effect2__), alpha=0.15) + 
  scale_linetype_discrete(labels = c("Negation","Lower Status","Higher Status")) + 
  scale_color_manual(labels = c("Negation","Lower Status","Higher Status"), values = wes_palette("Darjeeling1")) + 
  scale_fill_manual( 
                      labels=c("Negation","Lower Status","Higher Status"), values = wes_palette("Darjeeling1")) + 
  theme(
    legend.position = c(.08, .75),
    legend.justification = c("left", "bottom"),
    legend.box.just = "left",
    legend.margin = margin(6, 6, 6, 6)
    ) + theme(axis.text=element_text(size=12),
        axis.title=element_text(size=12,face="bold")) + theme(legend.text = element_text(size=12)) + theme(panel.border = element_rect(colour = "black", fill = NA, size =1)) + theme(legend.title = element_blank()) + theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank(),
panel.background = element_blank(), axis.line = element_line(colour = "black")) +
  xlab("Self-Evaluation") + ylab("Ingroup Classification")
SR.ING.COND.2.plot

Similarity-to-Self Ingroup Classification

WSR.ING.COND.2.p <- conditional_effects(WSR.ING.COND.2)#, method = "posterior_linpred")
WSR.ING.COND.2.p_ <- WSR.ING.COND.2.p$`WSR.Z:outgroup`
WSR.ING.COND.2.plot <-ggplot(WSR.ING.COND.2.p_, aes(effect1__, estimate__)) +  geom_line(aes(linetype=effect2__, color=effect2__)) + 
  geom_ribbon(aes(ymin=lower__, ymax=upper__, fill=effect2__), alpha=0.15) + 
  scale_linetype_discrete(labels = c("Negation","Lower Status","Higher Status")) + 
  scale_color_manual(labels = c("Negation","Lower Status","Higher Status"), values = wes_palette("Darjeeling1")) + 
  scale_fill_manual( 
                      labels=c("Negation","Lower Status","Higher Status"), values = wes_palette("Darjeeling1")) + 
  theme(
    legend.position = c(.08, .75),
    legend.justification = c("left", "bottom"),
    legend.box.just = "left",
    legend.margin = margin(6, 6, 6, 6)
    ) + theme(axis.text=element_text(size=12),
        axis.title=element_text(size=12,face="bold")) + theme(legend.text = element_text(size=12)) + theme(panel.border = element_rect(colour = "black", fill = NA, size =1)) + theme(legend.title = element_blank()) + theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank(),
panel.background = element_blank(), axis.line = element_line(colour = "black")) +
  xlab("Similarity-to-Self") + ylab("Ingroup Classification")
WSR.ING.COND.2.plot

Combine Plots

COMB.ING.COND.2.plot1 <- ggarrange(DES.ING.COND.2.plot, SR.ING.COND.2.plot, WSR.ING.COND.2.plot, labels=c("A","B","C"), nrow=1)
COMB.ING.COND.2.plot2 <- ggarrange(DES.ING.COND.2.plot, SR.ING.COND.2.plot, WSR.ING.COND.2.plot, labels=c("D","E","F"), nrow=1)

Metacontrast Ratio

MCS.SR.COND.2.p <- conditional_effects(MCS.SR.COND.2)#, method = "posterior_linpred")

Warning: Predictions are treated as continuous variables in 'conditional_effects' by default which is likely invalid for ordinal families. Please set 'categorical' to TRUE.Warning: Predictions are treated as continuous variables in 'conditional_effects' by default which is likely invalid for ordinal families. Please set 'categorical' to TRUE.Warning: Predictions are treated as continuous variables in 'conditional_effects' by default which is likely invalid for ordinal families. Please set 'categorical' to TRUE.

MCS.SR.COND.2.p_ <- MCS.SR.COND.2.p$`metaContrastS.Z:outgroup`
MCS.SR.COND.2.MCRplot <-ggplot(MCS.SR.COND.2.p_, aes(effect1__, estimate__)) +  geom_line(aes(linetype=effect2__, color=effect2__)) + 
  geom_ribbon(aes(ymin=lower__, ymax=upper__, fill=effect2__), alpha=0.15) + 
  scale_linetype_discrete(labels = c("Negation","Lower Status","Higher Status")) + 
  scale_color_manual(labels = c("Negation","Lower Status","Higher Status"), values = wes_palette("Darjeeling1")) + 
  scale_fill_manual( 
                      labels=c("Negation","Lower Status","Higher Status"), values = wes_palette("Darjeeling1")) + 
  theme(
    legend.position = c(.12, .7),
    legend.justification = c("left", "bottom"),
    legend.box.just = "left",
    legend.margin = margin(6, 6, 6, 6)
    ) + theme(axis.text=element_text(size=12),
        axis.title=element_text(size=12,face="bold")) + theme(legend.text = element_text(size=12)) + theme(panel.border = element_rect(colour = "black", fill = NA, size =1)) + theme(legend.title = element_blank()) + theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank(),
panel.background = element_blank(), axis.line = element_line(colour = "black")) +
  xlab("Metacontrast Ratio") + ylab("Self-Evaluation")
MCS.SR.COND.2.MCRplot

Summed Similarity to Outgroup Compared Across Conditions

ISOS.SR.COND.2.p <- conditional_effects(ISOS.SR.COND.2)#, method = "posterior_linpred")

Warning: Predictions are treated as continuous variables in 'conditional_effects' by default which is likely invalid for ordinal families. Please set 'categorical' to TRUE.Warning: Predictions are treated as continuous variables in 'conditional_effects' by default which is likely invalid for ordinal families. Please set 'categorical' to TRUE.Warning: Predictions are treated as continuous variables in 'conditional_effects' by default which is likely invalid for ordinal families. Please set 'categorical' to TRUE.Warning: Predictions are treated as continuous variables in 'conditional_effects' by default which is likely invalid for ordinal families. Please set 'categorical' to TRUE.Warning: Predictions are treated as continuous variables in 'conditional_effects' by default which is likely invalid for ordinal families. Please set 'categorical' to TRUE.

ISOS.SR.COND.2.p_ <- ISOS.SR.COND.2.p$`inGsimS.Z:outgroup`
ISOS.SR.COND.2.InGplot <-ggplot(ISOS.SR.COND.2.p_, aes(effect1__, estimate__)) +  geom_line(aes(linetype=effect2__, color=effect2__)) + 
  geom_ribbon(aes(ymin=lower__, ymax=upper__, fill=effect2__), alpha=0.15) + 
  scale_linetype_discrete(labels = c("Negation","Lower Status","Higher Status")) + 
  scale_color_manual(labels = c("Negation","Lower Status","Higher Status"), values = wes_palette("Darjeeling1")) + 
  scale_fill_manual( 
                      labels=c("Negation","Lower Status","Higher Status"), values = wes_palette("Darjeeling1")) + 
  theme(
    legend.position = c(.12, .7),
    legend.justification = c("left", "bottom"),
    legend.box.just = "left",
    legend.margin = margin(6, 6, 6, 6)
    ) + theme(axis.text=element_text(size=12),
        axis.title=element_text(size=12,face="bold")) + theme(legend.text = element_text(size=12)) + theme(panel.border = element_rect(colour = "black", fill = NA, size =1)) + theme(legend.title = element_blank()) + theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank(),
panel.background = element_blank(), axis.line = element_line(colour = "black")) +
  xlab("Similarity-to-Ingroup") + ylab("Self-Evaluation")
ISOS.SR.COND.2.InGplot

Summed Similarity to Outgroup Compared Across Conditions

ISOS.SR.COND.2.p <- conditional_effects(ISOS.SR.COND.2)#, method = "posterior_linpred")

Warning: Predictions are treated as continuous variables in 'conditional_effects' by default which is likely invalid for ordinal families. Please set 'categorical' to TRUE.Warning: Predictions are treated as continuous variables in 'conditional_effects' by default which is likely invalid for ordinal families. Please set 'categorical' to TRUE.Warning: Predictions are treated as continuous variables in 'conditional_effects' by default which is likely invalid for ordinal families. Please set 'categorical' to TRUE.Warning: Predictions are treated as continuous variables in 'conditional_effects' by default which is likely invalid for ordinal families. Please set 'categorical' to TRUE.Warning: Predictions are treated as continuous variables in 'conditional_effects' by default which is likely invalid for ordinal families. Please set 'categorical' to TRUE.

ISOS.SR.COND.2.p_ <- ISOS.SR.COND.2.p$`outGsimS.Z:outgroup`
ISOS.SR.COND.2.OutGplot <-ggplot(ISOS.SR.COND.2.p_, aes(effect1__, estimate__)) +  geom_line(aes(linetype=effect2__, color=effect2__)) + 
  geom_ribbon(aes(ymin=lower__, ymax=upper__, fill=effect2__), alpha=0.15) + 
  scale_linetype_discrete(labels = c("Negation","Lower Status","Higher Status")) + 
  scale_color_manual(labels = c("Negation","Lower Status","Higher Status"), values = wes_palette("Darjeeling1")) + 
  scale_fill_manual( 
                      labels=c("Negation","Lower Status","Higher Status"), values = wes_palette("Darjeeling1")) + 
  theme(
    legend.position = c(.12, .7),
    legend.justification = c("left", "bottom"),
    legend.box.just = "left",
    legend.margin = margin(6, 6, 6, 6)
    ) + theme(axis.text=element_text(size=12),
        axis.title=element_text(size=12,face="bold")) + theme(legend.text = element_text(size=12)) + theme(panel.border = element_rect(colour = "black", fill = NA, size =1)) + theme(legend.title = element_blank()) + theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank(),
panel.background = element_blank(), axis.line = element_line(colour = "black")) +
  xlab("Similarity-to-Outgroup") + ylab("Self-Evaluation")
ISOS.SR.COND.2.OutGplot

Study 3

Warmth

warmthDf3 <- pivot_longer(indDiffs3, cols=starts_with("Therm_"), names_to="Group", values_to="Warmth") %>% select(subID, Group, Warmth) %>% drop_na()
warmthDf3$Group <- gsub("Therm_1","White", warmthDf3$Group)
warmthDf3$Group <- gsub("Therm_2","Asian", warmthDf3$Group)
warmthDf3$Group <- gsub("Therm_4","Latino", warmthDf3$Group)

warmthDf3$Group <- as.factor(warmthDf3$Group)
warmthDf3$subID <- as.factor(warmthDf3$subID)

warmPlot3 <- ggplot(warmthDf3, aes(Group, Warmth)) + 
  ggdist::stat_halfeye(adjust = .5, width = .7, .width = 0, justification = -.2, point_colour = NA) + 
  geom_boxplot(width = .2, outlier.shape = NA) + 
  geom_jitter(width = .05, alpha = .15) + labs(y="Perceived Warmth",x="Racial Group") + jtools::theme_apa()

Status

indDiffs3$Status <- NULL
statusDf3 <- pivot_longer(indDiffs3, cols=ends_with("_Status"), names_to="Group", values_to="Status") %>% select(subID, Group, Status) %>% drop_na()
statusDf3$Group <- gsub("_Status","", statusDf3$Group)

statusDf3$Group <- as.factor(statusDf3$Group)
statusDf3$subID <- as.factor(statusDf3$subID)

statusPlot3 <- ggplot(statusDf3, aes(Group, Status)) + 
  ggdist::stat_halfeye(adjust = .5, width = .7, .width = 0, justification = -.2, point_colour = NA) + 
  geom_boxplot(width = .2, outlier.shape = NA) + 
  geom_jitter(width = .05, alpha = .3) + labs(y="Perceived Status",x="Racial Group") + jtools::theme_apa()

Combine Warmth and Status Plots

wsPlot3 <- warmPlot3 + statusPlot3

wsPlot3

Projection Rates

mDiffPlot3 <- ggplot(params3, aes(condition, m)) + 
  ggdist::stat_halfeye(adjust = .5, width = .7, .width = 0, justification = -.2, point_colour = NA) + 
  geom_boxplot(width = .2, outlier.shape = NA) + scale_x_discrete(labels=c("Majority","Minority")) +
  geom_jitter(width = .05, alpha = .15) + labs(y="Projection Rate",x="Racial Group") + jtools::theme_apa()

Bias Parameter

bDiffPlot3 <- ggplot(params3, aes(condition, bias)) + 
  ggdist::stat_halfeye(adjust = .5, width = .7, .width = 0, justification = -.2, point_colour = NA) + 
  geom_boxplot(width = .2, outlier.shape = NA) + scale_x_discrete(labels=c("Majority","Minority")) +
  geom_jitter(width = .05, alpha = .15) + labs(y="Bias Parameter",x="Racial Group") + jtools::theme_apa()

Combine Bias and Projection Plots

mbPlot3 <- mDiffPlot3 + bDiffPlot3

mbPlot3

Desirability Ingroup Classification

DES.ING.COND.3.p <- conditional_effects(DES.ING.COND.3)
DES.ING.COND.3.p_ <- DES.ING.COND.3.p$`desirability.Z:condition`
DES.ING.COND.3.plot <-ggplot(DES.ING.COND.3.p_, aes(effect1__, estimate__)) +  geom_line(aes(linetype=effect2__, color=effect2__)) + geom_ribbon(aes(ymin=lower__, ymax=upper__, fill=effect2__), alpha=0.15) + scale_linetype_discrete(labels = c("Majority","Minority")) + scale_color_manual(labels = c("Majority","Minority"), values = c("#44AA99","#332288")) + scale_fill_manual( 
                      labels=c("Majority","Minority"), values = c("#44AA99","#332288")) + theme(
    legend.position = c(.12, .7),
    legend.justification = c("left", "bottom"),
    legend.box.just = "left",
    legend.margin = margin(6, 6, 6, 6)
    ) + theme(axis.text=element_text(size=12),
        axis.title=element_text(size=12,face="bold")) + theme(legend.text = element_text(size=12)) + theme(panel.border = element_rect(colour = "black", fill = NA, size =1)) + theme(legend.title = element_blank()) + theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank(),
panel.background = element_blank(), axis.line = element_line(colour = "black")) +
  xlab("Desirability") + ylab("Ingroup Classification")
DES.ING.COND.3.plot

Self-Evaluation Ingroup Classification

SR.ING.COND.3.p <- conditional_effects(SR.ING.COND.3)
SR.ING.COND.3.p_ <- SR.ING.COND.3.p$`selfResp.Z:condition`
SR.ING.COND.3.plot <-ggplot(SR.ING.COND.3.p_, aes(effect1__, estimate__)) +  geom_line(aes(linetype=effect2__, color=effect2__)) + geom_ribbon(aes(ymin=lower__, ymax=upper__, fill=effect2__), alpha=0.15) + scale_linetype_discrete(labels = c("Majority","Minority")) + scale_color_manual(labels = c("Majority","Minority"), values = c("#44AA99","#332288")) + scale_fill_manual( 
                      labels=c("Majority","Minority"), values = c("#44AA99","#332288")) + theme(
    legend.position = c(.12, .7),
    legend.justification = c("left", "bottom"),
    legend.box.just = "left",
    legend.margin = margin(6, 6, 6, 6)
    ) + theme(axis.text=element_text(size=12),
        axis.title=element_text(size=12,face="bold")) + theme(legend.text = element_text(size=12)) + theme(panel.border = element_rect(colour = "black", fill = NA, size =1)) + theme(legend.title = element_blank()) + theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank(),
panel.background = element_blank(), axis.line = element_line(colour = "black")) +
  xlab("Self-Evaluation") + ylab("Ingroup Classification")
SR.ING.COND.3.plot

Similarity-to-Self Ingroup Classification

WSR.ING.COND.3.p <- conditional_effects(WSR.ING.COND.3)
WSR.ING.COND.3.p_ <- WSR.ING.COND.3.p$`WSR.Z:condition`
WSR.ING.COND.3.plot <-ggplot(WSR.ING.COND.3.p_, aes(effect1__, estimate__)) +  geom_line(aes(linetype=effect2__, color=effect2__)) + geom_ribbon(aes(ymin=lower__, ymax=upper__, fill=effect2__), alpha=0.15) + scale_linetype_discrete(labels = c("Majority","Minority")) + scale_color_manual(labels = c("Majority","Minority"), values = c("#44AA99","#332288")) + scale_fill_manual( 
                      labels=c("Majority","Minority"), values = c("#44AA99","#332288")) + theme(
    legend.position = c(.12, .7),
    legend.justification = c("left", "bottom"),
    legend.box.just = "left",
    legend.margin = margin(6, 6, 6, 6)
    ) + theme(axis.text=element_text(size=12),
        axis.title=element_text(size=12,face="bold")) + theme(legend.text = element_text(size=12)) + theme(panel.border = element_rect(colour = "black", fill = NA, size =1)) + theme(legend.title = element_blank()) + theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank(),
panel.background = element_blank(), axis.line = element_line(colour = "black")) +
  xlab("Similarity-to-Self") + ylab("Ingroup Classification")
WSR.ING.COND.3.plot

Combine Plots

COMB.ING.COND.3.plot1 <- ggarrange(DES.ING.COND.3.plot, SR.ING.COND.3.plot, WSR.ING.COND.3.plot, labels=c("A","B","C"), nrow=1)
COMB.ING.COND.3.plot3 <- ggarrange(DES.ING.COND.3.plot, SR.ING.COND.3.plot, WSR.ING.COND.3.plot, labels=c("G","H","I"), nrow=1)

Metacontrast Ratio

MCS.SR.COND.3.p <- conditional_effects(MCS.SR.COND.3)

Warning: Predictions are treated as continuous variables in 'conditional_effects' by default which is likely invalid for ordinal families. Please set 'categorical' to TRUE.Warning: Predictions are treated as continuous variables in 'conditional_effects' by default which is likely invalid for ordinal families. Please set 'categorical' to TRUE.Warning: Predictions are treated as continuous variables in 'conditional_effects' by default which is likely invalid for ordinal families. Please set 'categorical' to TRUE.

MCS.SR.COND.3.p_ <- MCS.SR.COND.3.p$`metaContrastS.Z:condition`
MCS.SR.COND.3.MCRplot <-ggplot(MCS.SR.COND.3.p_, aes(effect1__, estimate__)) +  geom_line(aes(linetype=effect2__, color=effect2__)) + geom_ribbon(aes(ymin=lower__, ymax=upper__, fill=effect2__), alpha=0.15) + scale_linetype_discrete(labels = c("Majority","Minority")) + scale_color_manual(labels = c("Majority","Minority"), values = c("#44AA99","#332288")) + scale_fill_manual( 
                      labels=c("Majority","Minority"), values = c("#44AA99","#332288")) + theme(
    legend.position = c(.12, .7),
    legend.justification = c("left", "bottom"),
    legend.box.just = "left",
    legend.margin = margin(6, 6, 6, 6)
    ) + theme(axis.text=element_text(size=12),
        axis.title=element_text(size=12,face="bold")) + theme(legend.text = element_text(size=12)) + theme(panel.border = element_rect(colour = "black", fill = NA, size =1)) + theme(legend.title = element_blank()) + theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank(),
panel.background = element_blank(), axis.line = element_line(colour = "black")) +
  xlab("Metacontrast Ratio") + ylab("Self-Evaluation")
MCS.SR.COND.3.MCRplot

Summed Similarity to Ingroup Compared Across Conditions

ISOS.SR.COND.3.p <- conditional_effects(ISOS.SR.COND.3)

Warning: Predictions are treated as continuous variables in 'conditional_effects' by default which is likely invalid for ordinal families. Please set 'categorical' to TRUE.Warning: Predictions are treated as continuous variables in 'conditional_effects' by default which is likely invalid for ordinal families. Please set 'categorical' to TRUE.Warning: Predictions are treated as continuous variables in 'conditional_effects' by default which is likely invalid for ordinal families. Please set 'categorical' to TRUE.Warning: Predictions are treated as continuous variables in 'conditional_effects' by default which is likely invalid for ordinal families. Please set 'categorical' to TRUE.Warning: Predictions are treated as continuous variables in 'conditional_effects' by default which is likely invalid for ordinal families. Please set 'categorical' to TRUE.

ISOS.SR.COND.3.p_ <- ISOS.SR.COND.3.p$`inGsimS.Z:condition`
ISOS.SR.COND.3.InGplot <-ggplot(ISOS.SR.COND.3.p_, aes(effect1__, estimate__)) +  geom_line(aes(linetype=effect2__, color=effect2__)) + geom_ribbon(aes(ymin=lower__, ymax=upper__, fill=effect2__), alpha=0.15) + scale_linetype_discrete(labels = c("Majority","Minority")) + scale_color_manual(labels = c("Majority","Minority"), values = c("#44AA99","#332288")) + scale_fill_manual( 
                      labels=c("Majority","Minority"), values = c("#44AA99","#332288")) + theme(
    legend.position = c(.12, .7),
    legend.justification = c("left", "bottom"),
    legend.box.just = "left",
    legend.margin = margin(6, 6, 6, 6)
    ) + theme(axis.text=element_text(size=12),
        axis.title=element_text(size=12,face="bold")) + theme(legend.text = element_text(size=12)) + theme(panel.border = element_rect(colour = "black", fill = NA, size =1)) + theme(legend.title = element_blank()) + theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank(),
panel.background = element_blank(), axis.line = element_line(colour = "black")) +
  xlab("Similarity-to-Ingroup") + ylab("Self-Evaluation")
ISOS.SR.COND.3.InGplot

Summed Similarity to Outgroup Compared Across Conditions

ISOS.SR.COND.3.p <- conditional_effects(ISOS.SR.COND.3)

Warning: Predictions are treated as continuous variables in 'conditional_effects' by default which is likely invalid for ordinal families. Please set 'categorical' to TRUE.Warning: Predictions are treated as continuous variables in 'conditional_effects' by default which is likely invalid for ordinal families. Please set 'categorical' to TRUE.Warning: Predictions are treated as continuous variables in 'conditional_effects' by default which is likely invalid for ordinal families. Please set 'categorical' to TRUE.Warning: Predictions are treated as continuous variables in 'conditional_effects' by default which is likely invalid for ordinal families. Please set 'categorical' to TRUE.Warning: Predictions are treated as continuous variables in 'conditional_effects' by default which is likely invalid for ordinal families. Please set 'categorical' to TRUE.

ISOS.SR.COND.3.p_ <- ISOS.SR.COND.3.p$`outGsimS.Z:condition`
ISOS.SR.COND.3.OutGplot <-ggplot(ISOS.SR.COND.3.p_, aes(effect1__, estimate__)) +  geom_line(aes(linetype=effect2__, color=effect2__)) + geom_ribbon(aes(ymin=lower__, ymax=upper__, fill=effect2__), alpha=0.15) + scale_linetype_discrete(labels = c("Majority","Minority")) + scale_color_manual(labels = c("Majority","Minority"), values = c("#44AA99","#332288")) + scale_fill_manual( 
                      labels=c("Majority","Minority"), values = c("#44AA99","#332288")) + theme(
    legend.position = c(.12, .7),
    legend.justification = c("left", "bottom"),
    legend.box.just = "left",
    legend.margin = margin(6, 6, 6, 6)
    ) + theme(axis.text=element_text(size=12),
        axis.title=element_text(size=12,face="bold")) + theme(legend.text = element_text(size=12)) + theme(panel.border = element_rect(colour = "black", fill = NA, size =1)) + theme(legend.title = element_blank()) + theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank(),
panel.background = element_blank(), axis.line = element_line(colour = "black")) +
  xlab("Similarity-to-Outgroup") + ylab("Self-Evaluation")
ISOS.SR.COND.3.OutGplot

Studies Combined

wsPlot23 <- wsPlot2 / wsPlot3

wsPlot23
ggsave(wsPlot23, file = "/Volumes/Research Project/Self-Anchoring/Plots/warmStatus23.png", dpi = 300, units = "in", width= 10, height = 12)

ggsave(wsPlot23, file = "/Volumes/Research Project/Self-Anchoring/Plots/warmStatus23.tiff", dpi = 300, units = "in", width= 10, height = 12)

mbPlot23 <- mbPlot2 / mbPlot3

mbPlot23
ggsave(mbPlot23, file = "/Volumes/Research Project/Self-Anchoring/Plots/projBias23.png", dpi = 300, units = "in", width= 12, height = 9)

ggsave(mbPlot23, file = "/Volumes/Research Project/Self-Anchoring/Plots/projBias23.tiff", dpi = 300, units = "in", width= 12, height = 9)

COMB.ING.ALL <- ggarrange(COMB.ING.1.plot1, COMB.ING.COND.2.plot2, COMB.ING.COND.3.plot3, nrow =3)

ggsave(COMB.ING.ALL, file = "/Volumes/Research Project/Self-Anchoring/Plots/testPlot.png", dpi = 300, units = "in", width= 11, height = 14)

ggsave(COMB.ING.ALL, file = "/Volumes/Research Project/Self-Anchoring/Plots/testPlot.tiff", dpi = 300, units = "in", width= 11, height = 14)

trainingPlots <- ggpubr::ggarrange(MCS.SR.COND.2.MCRplot, ISOS.SR.COND.2.InGplot, ISOS.SR.COND.2.OutGplot,
          MCS.SR.COND.3.MCRplot, ISOS.SR.COND.3.InGplot, ISOS.SR.COND.3.OutGplot,
          labels = c("A", "B", "C", "D", "E", "F"),
          nrow = 2, ncol = 3)

ggsave(trainingPlots, file = "/Volumes/Research Project/Self-Anchoring/Plots/trainPlot.png", dpi = 300, units = "in", width= 10, height = 9)

ggsave(trainingPlots, file = "/Volumes/Research Project/Self-Anchoring/Plots/trainPlot.tiff", dpi = 300, units = "in", width= 10, height = 9)

For Talk

Study-Specific Generalization Phase Plots

SR.ING.1.plot + SR.ING.COND.2.plot + SR.ING.COND.3.plot
ggsave(file = "~/Desktop/SR.ING.png", dpi = 300, units = "in", width= 16, height = 5)

DES.ING.1.plot + DES.ING.COND.2.plot + DES.ING.COND.3.plot
ggsave(file = "~/Desktop/DES.ING.png", dpi = 300, units = "in", width= 16, height = 5)

WSR.ING.1.plot + WSR.ING.COND.2.plot + WSR.ING.COND.3.plot
ggsave(file = "~/Desktop/WSR.ING.png", dpi = 300, units = "in", width= 16, height = 5)

Study-Specific Training Phase Plots

MCS.SR.1.MCRplot + MCS.SR.COND.2.MCRplot + MCS.SR.COND.3.MCRplot
ggsave(file = "~/Desktop/MCS.SR.png", dpi = 300, units = "in", width= 16, height = 5)

ISOS.SR.1.InGplot + ISOS.SR.COND.2.InGplot + ISOS.SR.COND.3.InGplot
ggsave(file = "~/Desktop/InG.SR.png", dpi = 300, units = "in", width= 16, height = 5)

ISOS.SR.1.OutGplot + ISOS.SR.COND.2.OutGplot + ISOS.SR.COND.3.OutGplot
ggsave(file = "~/Desktop/OutG.SR.png", dpi = 300, units = "in", width= 16, height = 5)

wsPlot2
ggsave(file = "~/Desktop/wsPlot2.png", dpi = 300, units = "in", width= 10, height = 5)

wsPlot3
ggsave(file = "~/Desktop/wsPlot3.png", dpi = 300, units = "in", width= 10, height = 5)